Manage the gen 2 Openflow connector lifecycle

This topic describes how to manage the gen 2 Openflow connector lifecycle: start and stop ingestion and remove a gen 2 connector after it is created. To monitor connector health, throughput, and ingestion status, use the Openflow Connectors Dashboard. For gen 1 connectors, follow the lifecycle guidance in each connector’s public setup topic. See Openflow gen 1 and gen 2 for how to tell gen 1 resources from gen 2.

These tasks apply no matter how the gen 2 connector was created: using the Configure a connector with the setup wizard, Configure a gen 2 connector with SQL, or other supported SQL/API automation.

Start and stop data movement

After a gen 2 connector is installed on a runtime, start it to begin reading from the source and writing to Snowflake. Use Start from the connector’s menu on the Installed Connectors tab.

Use Stop when you need to pause ingestion—for example before maintenance, upgrades described in your connector’s documentation, or before removal. Stopping leaves the connector installed but idle.

Tip

Some connectors retain external resources while stopped (for example, database replication slots). Do not leave connectors stopped for long periods on busy sources unless you understand the impact; see your connector’s setup or maintenance topic.

For gen 2 connectors, use Start and Stop from the connector menu on Installed Connectors, or run ALTER OPENFLOW CONNECTOR ... START or STOP with SQL. Do not use the canvas for configuration or processor-level start/stop. See ALTER OPENFLOW CONNECTOR for syntax and wait functions.

Remove a connector

Gen 2 connector removal follows stop → terminate → drop. TERMINATE drains in-flight data before removal. Complete each step before starting the next.

In UI-driven workflows, wait for each step to finish before starting the next. In scripts, call SYSTEM$WAIT_FOR_STABLE_OPENFLOW_CONNECTORS after asynchronous ALTER commands. See SYSTEM$WAIT_FOR_STABLE_OPENFLOW_CONNECTORS.

Remove a connector (UI)

  1. From the Installed Connectors tab, open the connector menu and select Stop.
  2. From the connector menu on Installed Connectors, select Delete (terminates the connector and drains in-flight data).
  3. From the same menu, select Drop.

Remove a connector (SQL)

Delete in the UI corresponds to ALTER OPENFLOW CONNECTOR ... TERMINATE; Drop corresponds to DROP OPENFLOW CONNECTOR. DROP requires OWNERSHIP on the connector.

ALTER OPENFLOW CONNECTOR my_db.my_schema.my_connector STOP;
SELECT SYSTEM$WAIT_FOR_STABLE_OPENFLOW_CONNECTORS(600, 'my_db.my_schema.my_connector');

ALTER OPENFLOW CONNECTOR my_db.my_schema.my_connector TERMINATE;
SELECT SYSTEM$WAIT_FOR_STABLE_OPENFLOW_CONNECTORS(600, 'my_db.my_schema.my_connector');

DROP OPENFLOW CONNECTOR my_db.my_schema.my_connector;

For full command syntax, privileges, and additional ALTER options, see ALTER OPENFLOW CONNECTOR.

Caution

Delete, Drop, TERMINATE, and DROP OPENFLOW CONNECTOR are irreversible for Openflow entities. Snowflake does not support undrop for these objects. These steps do not remove destination tables or external resources (such as PostgreSQL replication slots). Read confirmation dialogs carefully.

For source-specific cleanup after removal (for example PostgreSQL replication slots), see Stop or delete the connector in Set up the Openflow Connector for PostgreSQL and analogous sections for other connectors.